Graduation Year
2021
Document Type
Dissertation
Degree
Ph.D.
Degree Name
Doctor of Philosophy (Ph.D.)
Degree Granting Department
Computer Science and Engineering
Major Professor
Srinivas Katkoori, Ph.D.
Committee Member
Swaroop Ghosh, Ph.D.
Committee Member
Jay Ligatti, Ph.D.
Committee Member
Andrew Hoff, Ph.D.
Committee Member
Brendan Nagle, Ph.D.
Keywords
Cryptography, Hardware Oriented Security and Trust, Intellectual Property Protection
Abstract
Over the last two decades or so, VLSI hardware is increasingly subject to sophisticated attacks on both the supply chain and design fronts. There is no explicit trust that the manufacturers/providers are not producing counterfeit designs or that cryptographic algorithms we know to be secure in software are also secure in hardware. The novelty and key contributions of this work are as follows: 1) a continually refined method for Intellectual Property (IP) Protection that provides an approach for verification of IP ownership, 2) demonstrate how to break the PRESENT-80 cryptographic algorithm with significantly limited resources, and 3) provide a multitude of hardware based countermeasures to counter such attacks. First, in order to thwart intellectual property theft, the proposed state encoding based watermarking method and the mapping algorithm outperforms the prior techniques. We propose a hybrid genetic algorithm, dubbed the Darwinian Genetic Algorithm, for efficiently solving the difficult sub-graph matching problem. As a result, we outperformed prior work maximally 20-30\% and on average 1-12\% when considering the post-synthesis watermarked designs in terms of literals, area, and delay. Second, we demonstrate for the first time ever how the lightweight cryptographic algorithm PRESENT-80 can be broken via a Differential Plaintext Attack with significantly limited resources. Lastly, to prevent such attacks, we present a series of countermeasures to not only PRESENT-80, but for all substitution-permutation network ciphers, by inducing non-static behavior. We present novel interconnection network primitives, dynamic routing networks, and ultimately modify round invariant items to round based variants that necessitate decision making for an attacker.
Scholar Commons Citation
Lewandowski, Matthew Dean, "Secure VLSI Hardware Design Against Intellectual Property (IP) Theft and Cryptographic Vulnerabilities" (2021). USF Tampa Graduate Theses and Dissertations.
https://digitalcommons.usf.edu/etd/9166
